[Jarrod]

Can an Agent persona go into a Host while the decker's persona stays on the grid? If the decker's persona is in silent mode, are the Agents in silent mode? Can an Agent be in silent mode while the decker's persona is not?

[fseperent]

You can only have a single agent or persona working from a single device at a time.
Cannot say for sure if an agent can run in silent mode.

[Herr Brackhaus]

Can an Agent persona go into a Host while the decker's persona stays on the grid?

An Agent is an autonomous program (SR5 page 246), and entering a host is a Matrix Action (SR5 page 239) that an Agent program is capable of performing. The only requirement is that the Agent must have at least one MARK on the host, and MARKs can't be shared, so it would have to obtain said MARK on it's own. This is potentially problematic, especially against higher rating hosts, but not impossible, so the answer to this question is "Yes, absolutely."

If the decker's persona is in silent mode, are the Agents in silent mode? Can an Agent be in silent mode while the decker's persona is not?

Not necessarily, and absolutely, respectively.

First of all, a decker can run an Agent program on a commlink and his own persona on a deck thus separating the two devices entirely for the purposes of Running Silently.

Secondly, when someone uses a device to access the Matrix that device icon is subsumed by the Persona icon (SR5 page 235), but an Agent running on the same device has it's own icon in turn. Since Running Silently can be toggled for on a per device or persona basis (SR5 page 235) you can absolutely have an Agent run silently while the decker doesn't, or vice versa.

Finally, fseperent's comment about having a single agent or persona from a single device is not entirely correct; while it's true that any one person can only have a single persona at a time (SR5 p235) it may or may not be possible to run multiple agents on a single device, depending on your interpretation of the rules, and it's definitively possible to run two personas on a single device by virtue of having an agent and a person using the same device.

The multiple agent on a single device topic has come up several times, and it really boils down to how one interprets the statement that agents are programs, and one cannot run the same program more than once on a single device to gain the benefit twice. The argument usually comes down to whether two Agent programs are the same as two Edit programs, for example, or if a Rating 3 agent program is the same as a Rating 4 agent program.

All I can say is that at my table we allow multiple agent programs, even of the same rating, to run simultaneously on a single device. The significant up-front investment cost (agents are expensive, after all) along with the increased risk of both exposure and damage (each additional agent is one more icon that can be spotted and/or attacked, and all damage is taken by the device they're running on) balance the benefits of running multiple agents. This is obviously true for my table and it doesn't have to be for others, and the rules here are ambiguous enough that multiple interpretations are possible.

[fseperent]

It was late so might of been thinking of 4th's rules.
There is a need for caution though, for every persona/agent running at once beyond the 1st, you suffer a -5 dice pool except when targeting other personas or agents on the same device.

[Herr Brackhaus]

That's an AI specific rule (Data Trails page 153) that I would be extremely hesitant to use in any other case, fseperent, as that would otherwise apply to an Agent and a decker having their personas on the same device. In essence, that rule is listed in a section that details how an AI can effectively perform a hostile takeover of a device.

[fseperent]

Sorry, mind's not working.

[DigitalZombie]

First of all, a decker can run an Agent program on a commlink and his own persona on a deck thus separating the two devices entirely for the purposes of Running Silently.

Hm Im not sure that you can run 2 links/decks at the same time. Im pretty sure the decker needs to have his team mate carry the deck and "run" it. And no, I dont know how that is explained in-game, it just happens auto-magically.

[Herr Brackhaus]

DigitalZombie
The only requirement by RAW is that a single person can only ever have a single persona. There is nothing in RAW that prohibits a single person from running multiple devices, whether they are capable of forming personas or not.

You can even slave other devices to your PAN, so not only is running multiple devices possible it's encouraged to protect them.

An agent running on a device without an active persona subsumes the device icon just like a normal person would; away from the books at the moment but that reference is in either the Persona, PAN or Agent sections referenced above.

Besides, if the Decker needed to have a buddy run his agent device, how would his team mate run his own link?

How would any character run any agent program on a cheap decker-in-a-pocket setup while also using their own high rating commlink?

From an in-game point of view this is entirely consistent; just like any other device, a device hosting an Agent program is still a device and can be operated by someone who has formed their persona elsewhere. Just like a Rigger forming his persona on an RCC can still "use" drones with a pilot software, such a character could run a low-rating deck with an Agent program on it to look for hostile MARKs.

[bull30548]

So this sort of answers a question I had.  I was thinking of pulling something similar with a few cheap decks and link them all to a primary deck.  Then each one has an agent program running on it and I can essentially be a technomancer with some decent sprites assisting me.  The drawback is that I would rule it that even if the agents are all assisting me that my OS score should spike faster by doing this.  However, to be honest the way RAW is that an agent assisting doesn't generate OS because it isn't outright attacking it is lending it's support.   So my question is if I link say 3 cheap decks (or even certain commlinks) together and using them to give me an assist.  This is for a one shot big brute force attack not to be used all the time though I think it could be useful.

[Herr Brackhaus]

Hard to tell what RAW is on this because it's so vague, but it certainly seems that the intention is that OS is generated when an illegal action is performed.

At our table, we share OS between team members when an action is performed, so both the decker and his agents as well as the technomancer and his sprites earn an equal amount of OS. This next part is almost certainly not RAW, but we also allow the sharing, for lack of a better word, of MARKs; in other words, when the decker is assisted by an Agent to hack a mark on a host and is successful,  the decker and agent can both enter the host with the single MARK gained. No need to make the hacking game take even longer by requiring twice the amount of rolls, after all...

[Kaeylum]

So this sort of answers a question I had.  I was thinking of pulling something similar with a few cheap decks and link them all to a primary deck.  Then each one has an agent program running on it and I can essentially be a technomancer with some decent sprites assisting me.  The drawback is that I would rule it that even if the agents are all assisting me that my OS score should spike faster by doing this.  However, to be honest the way RAW is that an agent assisting doesn't generate OS because it isn't outright attacking it is lending it's support.   So my question is if I link say 3 cheap decks (or even certain commlinks) together and using them to give me an assist.  This is for a one shot big brute force attack not to be used all the time though I think it could be useful.

I don't think there's anything in the RAW that would prevent multiple agents from helping you, except that they each take a program slot on the device, and team work tests bonuses max out at whatever your rating in the skill are.  And as far as the agents generating overwatch, if they're performing illegal actions, they would, but it would be attributed to their persona.  Unfortunately their persona, if I understand it correctly is tied to your deck, so in the event of convergence, GOD would still know where you were.

[Kuirem]

Agents are Programs. You can't run more than one program of the same type on your deck. So you can't run multiple agent on your deck.

[Coyote]

Agents are Programs. You can't run more than one program of the same type on your deck. So you can't run multiple agent on your deck.

There is a reasonable debate whether a Rating 2 Agent is a different program from a Rating 3 Agent.

[Herr Brackhaus]

Agents are Programs. You can't run more than one program of the same type on your deck. So you can't run multiple agent on your deck.

I'd argue that since RAW states that agents have their own persona this means that any two agent programs cannot be identical. Thus, each agent program is a unique program in and of itself, and the usual limitation of not being able to run two copies of the "same program" on a device doesn't apply.

Mostly, I feel this isn't a problem because all you're doing is increasing your attack surface. Since all agents would share the same condition monitor and have worse dice pools than most dedicated hackers, you'd be risking almost casual destruction for a few dice. But that's a personal opinion.

[Kuirem]

That feels like a huge twist just to allow it. RAW doesn't even say "same program" but "program of the same type". Is there even an argument that Agent 2 is the same type as Agent 3? Knowing that they have the exact same functionalities? Increasing the attack surface don't feel like a big deal especially if you just use all your agents to just run Matrix Perception. Imagine 4 Agent Rating 4 running Matrix Perception, it's impossible to hide from them and you just desactivate them when you switch to fight. Easily exploitable.